Profile of Women Victims of Gender Violence in Rural Settings: Mental Health and Risk Perception

Abstract

Gender-based violence constitutes a major public health and social concern, with particularly complex implications in structurally vulnerable contexts such as rural settings. However, empirical evidence regarding the specific profile and risk perception of women experiencing gender-based violence in small municipalities remains limited. The aim of this study was to analyze the sociodemographic characteristics of women victims of gender-based violence residing in small rural municipalities and to examine their associations with mental health indicators and perceived risk of future violence. The sample comprised 30 women receiving support at a Specialized Care Centre for Victims of Gender-Based Violence (CAVI) serving three small municipalities in the Vega Media region (Region of Murcia, Spain). Standardized measures of depression, anxiety, and stress were administered, together with an assessment of perceived risk. The findings suggest a specific sociodemographic profile characterized by moderate levels of depression, anxiety, and stress symptoms and generally low perceived risk. Women without children reported higher levels of psychological distress and perceived risk than those with children, although these differences should be interpreted with caution given the sample size. Overall, these findings provide preliminary insights into the characteristics and risk perception of women experiencing gender-based violence in rural settings and highlight the need for context-sensitive prevention and intervention strategies.

1. Introduction

Violence against women, particularly in the form of gender-based violence perpetrated by intimate partners, remains one of the most pervasive violations of human rights worldwide and a significant public health challenge. Rooted in structural gender inequality, it reflects enduring power imbalances that place women at heightened risk of physical, psychological and social harm [1,2,3]. Unlike other forms of violence, gender-based violence is embedded within unequal power relations and produces distinct psychological, social and health-related consequences. Despite legislative advances and institutional efforts, it remains a persistent and deeply concerning social problem [4,5,6,7,8].

Although gender-based violence affects women across all sociodemographic groups [9], research has identified specific demographic characteristics associated with increased vulnerability [10,11,12,13,14,15]. Riggs et al. [16] identified family stress, low socioeconomic status and unemployment as potential risk factors. In a study of 1081 officially reported cases, Echeburúa et al. [17] found that victims were typically younger than perpetrators, with a mean age of approximately 35 years. A substantial proportion were immigrant women (27%), frequently unemployed or engaged in low-skilled occupations, with limited educational attainment and low or lower-middle socioeconomic status.

Exposure to gender-based violence is consistently associated with a higher prevalence of mental health disorders compared with women who have not experienced such violence [18,19]. It has been estimated that approximately 60% of women exposed to intimate partner violence (IPV) develop moderate to severe psychological disorders [20,21]. Depression and post-traumatic stress disorder are among the most frequently reported conditions [22,23,24,25,26], although anxiety disorders [27,28,29], substance misuse [30,31] and suicidal ideation [32,33] are also highly prevalent.

From a psychological perspective, women exposed to IPV frequently report feelings of helplessness, diminished self-esteem, and negative self-appraisal, which may impair their capacity to cope effectively with the abusive situation [34]. Early theoretical accounts, such as the “Battered Woman Syndrome” proposed by Walker [35], conceptualized these responses within the framework of learned helplessness. However, this perspective has been subject to substantial criticism. Several authors have argued that BWS may pathologize survivors’ responses, portraying them as psychologically impaired rather than as individuals adapting to constrained and coercive circumstances [27,36,37,38].

More recent approaches emphasize the role of coercive control, understood as a pattern of domination that extends beyond physical violence and includes psychological, economic, and social constraints. In this regard, Stark [38] conceptualizes intimate partner violence as a form of ongoing entrapment, highlighting the structural and contextual factors that shape women’s responses and risk perception. Within this framework, cognitive and behavioral responses, such as minimization or normalization of violence, may be understood not only as defensive mechanisms but also as adaptive strategies in contexts of constrained agency [36,37].

Repeated victimization within intimate relationships may therefore lead to cumulative psychological burden and increased vulnerability to mental health disorders [38].

Risk perception represents a key factor in understanding continued exposure to violence. However, evidence suggests that women subjected to IPV may underestimate their risk of future victimization, potentially limiting the adoption of protective behaviors [39]. Consequently, structured risk assessment protocols have been developed internationally to provide systematic tools for evaluating and predicting the likelihood of future violence [40,41,42,43].

From an integrative and probabilistic perspective, violent behavior results from the interaction between risk and protective factors. The presence of a risk factor does not deterministically lead to violence, nor does its absence guarantee protection. Rather, risk factors operate as probabilistic enablers within a multifactorial causal framework [44].

The contextual environment in which violence occurs plays a crucial role in shaping vulnerability and access to support. Rural settings are characterized by structural constraints such as population ageing, limited employment opportunities for women, entrenched gender norms, increased social control and restricted access to specialized services. A report commissioned by the Spanish Ministry of Health, Social Services and Equality [45], together with more recent institutional analyses by the Government Delegation against Gender Violence [11], highlights the increased vulnerability of women residing in small municipalities, where structural barriers and limited institutional resources may discourage formal reporting and reduce the perceived feasibility of leaving abusive relationships.

The present study was conducted in the Vega Media region (Region of Murcia, Spain), an area composed of small municipalities with rural and semi-rural characteristics. This territorial context is characterized by relatively small populations, limited access to specialized services, and strong community ties, which may influence help-seeking behavior, social support dynamics, and risk perception among women experiencing gender-based violence. Understanding gender-based violence within this specific geographical and social context is therefore particularly relevant.

Given the specific characteristics of rural environments, further research is needed to better understand the sociodemographic profile, mental health status, and perceived risk among women experiencing gender-based violence in these settings. Such knowledge may inform the development of context-sensitive prevention and intervention strategies tailored to rural areas.

2. Study Aims

This study aims to examine gender-based violence among women residing in small rural municipalities. Specifically, the objectives are: (1) to describe the sociodemographic profile of women victims of gender-based violence; (2) to examine the relationships between sociodemographic factors, mental health indicators (depression, anxiety, and stress), and perceived risk of future violence; (3) to analyze the association between type of abuse and perceived risk of future violence.

3. Results

Normality was assessed using the Shapiro–Wilk test. Most variables showed distributions consistent with normality (p > 0.05), supporting the use of parametric analyses. Only the variable “duration” significantly deviated from normality (p < 0.05); therefore, results involving this variable should be interpreted with caution.

In relation to the first objective, the most relevant sociodemographic characteristics of women exposed to gender-based violence (see

Table 1. Sociodemographic characteristics of the sample.

Age	30% are young (17–34) 46.7% are adults (35–46) 23.3% are adults-seniors (47–75)
Nationality	87.6% are Spanish 13.3% are not Spanish
Educational level	96.6% have some studyeducation 3.3% are uneducated
Employment level	46.7% are active 46.7% are not active 3.3% study 3.3% oin sick leave
Economic situation	83.3% financially independent 13.3% financially dependent 3.3% Basic Reinsertion Income
Children	76.7% have children 20% have not children 3.3% are pregnantcy
Family support	80% have family support 20% have not family support
Current relationship with the aggressor	43.2% have an ex-partner 16.7% have a husband 26.7% have an ex-husband 13.3% are in the separation process
Duration of the relationship with the aggressor	70% from 1 to 15 years 20% from 15 to 30 years 10% more than 30 years
Type of abuse	66.7% physical mistreatment 80% psychological mistreatment 13.3% sexual mistreatment 23.3% economic mistreatment
Complaint	80% have reported 20% have not reported

) indicate a profile predominantly composed of women aged between 35 and 46 years (46.7%), of Spanish nationality (87.6%), with primary education (63.3%), and either employed or unemployed in equal proportions (46.7% in each case). Most participants reported being financially independent (83.3%), having children (76.7%), and receiving family support (80%).

Regarding their relationship with the perpetrator, 43.3% indicated that he was their ex-boyfriend or former partner. The mean duration of the relationship was 13.03 years, with 70% reporting a duration between 1 and 15 years. Psychological abuse was the most frequently reported type of violence (80%), followed by physical abuse (66.7%). Additionally, 80% of the women had filed a formal complaint.

In relation to the second objective,

Table 2. Mean scores of the sample on the DASS-21 scales.

DASS-21
Depression	Anxiety	Stress	Total
10.73	8.6	10.63	29.97

presents the mean scores obtained on mental health indicators (depression, anxiety, and stress), and

Table 3. Perception of the risk of violence against women (in percentages).

RVD-BCN
Low Risk	Medium Risk	High Risk
60.7	21.4	17.9

shows the distribution of participants according to their level of perceived risk of future violence based on the RVD-BCN.

These results indicate that participants demonstrated moderate levels of depressive, anxiety, and stress symptoms according to the DASS-21 interpretative cut-off scores. The mean total DASS-21 score was 29.97, which falls within the moderate severity range [21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42].

Regarding perceived risk of future violence, the majority of women (60.7%) were classified as low risk, whereas 39.3% were classified as medium or high risk.

Table 4. Pearson correlation analysis.

		Age	Total Symptomatology	Perception of Total Risk
Age	Pearson correlation	1	−0.17	−0.23
	Sig. (bilateral)		0.37	0.22
	N	30	30	28
Total symptomatology	Pearson correlation	−0.17	1	0.15
	Sig. (bilateral)	0.37		0.43
	N	30	30	28
Perception of total risk	Pearson correlation	−0.23	0.15	1
	Sig. (bilateral)	0.22	0.43
	N	28	28	28

presents the correlations between age, clinical symptomatology, and perceived risk of future violence. No statistically significant associations were found.

To analyze the relationship between the presence of children, mental health indicators, and perceived risk of future violence, independent samples t-tests were conducted. The independent variable was whether the women had children, and the dependent variables were depressive symptoms, anxiety, stress, total symptomatology, and perceived risk.

Table 5. T-test for independent samples.

		F	Sig.	T	gl	Sig. (Bilateral)	Diff. Mean
Total symptomatology	Equal variances are assumed	0.44	0.52	−2.19	27	0.04	−11.98
	Equal variances are not assumed			−1.91	6.76	0.09	−11.98
Depressive symptomatology	Equal variances are assumed	0.61	0.44	−2.47	27	0.02	−5.51
	Equal variances are not assumed			−2.78	9.33	0.02	−5.51
Anxious symptomatology	Equal variances are assumed	2.42	0.13	−1.04	27	0.31	−2.45
	Equal variances are not assumed			−0.76	6.02	0.46	−2.45
Stress symptomatology	Equal variances are assumed	0.04	0.85	−2.07	27	0.05	−4.00
	Equal variances are not assumed			−2.06	7.73	0.07	−4.00

shows statistically significant differences in total clinical symptoms, depressive symptoms, and stress; however, these findings should be interpreted with caution given the small sample size and the exploratory nature of the analyses. Women without children tended to report higher levels of symptomatology and perceived risk.

Regarding the third objective, to determine the relationship between the type of abuse (physical, psychological, sexual, and economic) and the level of perceived risk of future violence,

Table 6. Multiple regression analysis. (a) Summary of the model. (b) ANOVA ^b. (c) Coefficients ^d. Unstandardized coefficients.

(a)
Model	R	R Squared		R Squared Adjusted	Standard Error
1	0.57 a	0.33		0.21	2.60
(b)
Model		Sum of Squares	gl	Root Mean Square	F	Sig.
1	Regression	77.41	4	19.35	2.85	0.04 c
	Residual	155.84	23	6.77
	Total	233.25	27
(c)
Model		B	Standard Error	Standardized Beta Coefficients	t	Sig.
1	(Constant)	7.28	1.67		4.35	0.00
	Physical mistreatment	1.71	1.20	0.28	1.43	0.16
	Psychological mistreatment	−2.05	1.47	−0.27	−1.39	0.17
	Sexual abuse	−0.06	1.60	−0.00	−0.04	0.96
	Economic mistreatment	−1.76	1.20	−0.26	−1.46	0.15

^a: Predictors: (Constant) Economic mistreatment, physical mistreatment, sexual abuse, psychological mistreatment. ^b: Dependent variable: Total risk perception. ^c: Predictors: (Constant) Economic mistreatment, physical mistreatment, sexual abuse, psychological mistreatment. ^d: Dependent variable: Total risk perception.

shows that these predictors were statistically associated with perceived risk, explaining 21% of its variance; however, these results should be interpreted with caution given the sample size and the lack of statistically significant individual predictors.

4. Discussion

The present study aimed to examine gender-based violence among women residing in small rural municipalities. Specifically, it sought to describe the sociodemographic characteristics of women exposed to gender-based violence, to analyze the associations between these sociodemographic variables, mental health indicators (depression, anxiety, and stress), and perceived risk of future violence, and to explore the relationship between types of abuse and perceived risk of future violence.

With regard to mental health, participants reported, on average, moderate levels of depressive, anxiety and stress symptoms. This pattern is broadly consistent with previous evidence indicating that psychological sequelae among women exposed to gender-based violence can range from moderate to severe [3,46,47], although the present findings should be interpreted with caution. In the present sample, depressive symptoms showed the highest mean values, followed closely by stress. This aligns with research reporting a predominance of depressive symptomatology and trauma-related distress among women exposed to intimate partner violence (IPV) [22,23,25,26,48,49,50,51]. Nevertheless, symptom levels largely fell within the moderate range. One plausible explanation is the relatively high prevalence of family support reported in this study (80%), as social support has been repeatedly identified as a key protective factor that may buffer the psychological impact of victimization [52,53,54].

Regarding perceived risk, most participants (60.7%) were classified as low risk, which is broadly consistent with findings by González and Mora [55], who described a victim profile characterized by low-to-moderate perceived risk. It should be noted that the risk assessment data were available for 28 participants, as two women did not complete the questionnaire due to specific circumstances (the aggressor’s death and residence abroad). The predominance of low perceived risk may also be partly explained by the high levels of family support, which has been described as a protective resource potentially reducing continued exposure to IPV [18]. In addition, it is possible that the support and safety planning provided by the specialized center (CAVI) contributed to lower perceived risk; however, this interpretation should be treated cautiously and would require direct measurement of service-related variables.

No statistically significant associations were observed between age, clinical symptomatology and perceived risk. In this sample, being younger or older did not appear to be related to higher levels of distress or to greater perceived risk. This finding may reflect the fact that gender-based violence occurs across diverse age groups and sociodemographic contexts [3,56,57], and that its psychological consequences can be substantial irrespective of age. This interpretation is consistent with Sarasua et al. [58], who reported comparable anxiety levels across younger and older women. Similarly, the absence of an association between age and risk perception aligns with previous evidence suggesting that both younger and older women may experience difficulties identifying violence or appraising its severity [59,60].

The relationship between having children, symptomatology and perceived risk suggested a potential pattern: women without children reported higher total symptomatology, particularly depressive and stress symptoms, as well as higher perceived risk. A potential explanation for lower symptom reporting among women with children may be linked to maternal caregiving processes and the emotional bond established within the mother–child relationship [61,62,63,64]. From this perspective, the need to maintain caregiving functioning and protect the child may encourage emotional suppression, prioritization of the child’s needs, and a shift towards a relational “we” rather than an individual “I”, which could influence symptom reporting and coping strategies. This interpretation remains tentative and should be explored in future research using designs capable of assessing caregiving-related mechanisms directly.

A related finding was that women with children showed lower perceived risk. This may appear counterintuitive, given that the presence of children can increase vulnerability and complicate help-seeking and decision-making in abusive relationships [41]. Prior research has shown that some women may prefer to maintain the relationship to preserve the child’s contact with the father or due to fears regarding custody and child removal [65]. Such concerns may contribute to minimization of risk or to a reduced willingness to acknowledge risk explicitly. Conversely, women without children may perceive fewer barriers to leaving and may therefore appraise risk more directly, which could partially account for the higher perceived risk observed in this group.

With respect to the relationship between types of abuse and risk perception, the set of abuse indicators (physical, psychological, sexual and economic) was statistically associated with perceived risk, explaining 21% of the variance. This may suggest a potential relationship between the accumulation and severity of abuse experiences and women’s appraisal of danger; however, this interpretation should be considered provisional given the limitations. Conceptually, this is consistent with the view that multiple forms of victimization simultaneously threaten physical integrity, psychological wellbeing, sexual autonomy and financial independence, thereby intensifying perceived threat and the perceived likelihood of future harm.

Despite the contribution of the present study to the understanding of gender-based violence in rural contexts, several limitations should be acknowledged. First, the small sample size limits statistical power and restricts the generalizability of the findings. Nonetheless, the study addresses a vulnerable and hard-to-reach population, and the observed patterns may provide preliminary insights that could inform future research. Another limitation relates to the geographical scope of the study. The sample was drawn from a specialised care centre serving three municipalities in the Vega Media region (Murcia, Spain), and therefore the findings should be interpreted within this specific regional and rural context. The results cannot be generalized to all rural populations or to other national or international contexts without caution, as social, cultural, and institutional factors may differ across regions and countries. Second, the sample was drawn from a single specialized center, which may limit external validity, although it reflects the real-world context of service users in the region. Third, heterogeneity in age and other sociodemographic characteristics may have reduced the ability to detect associations. Future research would benefit from larger, multi-site samples and from designs that allow stratification by key variables (e.g., age, nationality, relationship status, and type of abuse), alongside more detailed measurement of service-related factors (e.g., duration of support, safety planning, and resource utilization).

5. Materials and Methods

5.1. Design

A cross-sectional descriptive and correlational study was conducted, following the methodological guidelines proposed by Ato et al. [66] who classify research designs in the behavioral sciences according to their objective (descriptive, comparative, or explanatory) and temporal structure. In line with this framework, the present study is explicitly defined as a descriptive and correlational design of a cross-sectional nature, aimed at examining relationships between variables without establishing causal inferences.

5.2. Participants

The sample was incidental and comprised 30 women exposed to gender-based violence attending a Specialized Care Centre for Women Victims of Gender Violence (CAVI) serving three small municipalities in southeastern Spain. Participants’ ages ranged from 17 to 75 years (M = 39.80, SD = 12.61).

5.3. Instruments

Sociodemographic information was obtained using the questionnaire routinely administered at the CAVI during the initial interviews with the women. Once completed by the professional on the basis of the information provided by the participants, the relevant variables for this study were extracted and compiled in a Word document. The variables included age, nationality, educational level, employment status, economic situation, presence of children, family support, current relationship with the aggressor, duration of the relationship, type of abuse, and whether a formal complaint had been filed.

Mental health indicators were assessed using the Spanish adaptation of the Depression, Anxiety and Stress Scale (DASS-21) developed by Antúnez and Vinet [67]. The instrument consists of 21 items grouped into three subscales (seven items each) assessing Depression, Anxiety, and Stress. Responses are recorded on a four-point Likert scale ranging from 0 (“Did not apply to me at all”) to 3 (“Applied to me very much, or most of the time”). Items refer to symptoms experienced during the previous seven days in the context of the maltreatment situation. Subscale scores are obtained by summing the corresponding items. The instrument demonstrates good internal consistency, with Cronbach’s α coefficients of 0.85 for Depression, 0.83 for Stress, and 0.73 for Anxiety, and 0.91 for the total scale.

Perceived risk of future violence was assessed using the Protocol for the Assessment of the Risk of Violence against Women by their Partners or Ex-Partners (RVD-BCN) developed by Álvarez et al. [41]. This instrument was created within the framework of the Barcelona Circuit against Violence against Women to assess the risk of serious violence by a partner or former partner in the short term (three to six months). The RVD-BCN comprises 16 items (risk factors) organized into five categories: history of violent behavior by the aggressor; threats and/or severe abuse towards the woman; aggravating circumstances; vulnerability factors of the woman; and the woman’s perception of risk. Each item is rated as “Yes”, “No”, or “Unknown”, and the total number of affirmative responses corresponds to a risk classification according to predefined cut-off points: low risk (1–7 points), medium risk (8–9 points), and high risk (10–16 points). The RVD-BCN does not automatically determine the level of risk; rather, it supports professional judgement. Responsibility for risk classification rests with the professional conducting the assessment, and the protocol aims to enhance precision, transparency, and decision-making in subsequent intervention planning [41]. The instrument demonstrates acceptable reliability, with a Cronbach’s α of 0.72 and interobserver reliability (Kendall’s Tau) of 0.73.

5.4. Procedure

The study was conducted in collaboration with the Specialised Care Centre for Women Victims of Gender Violence (CAVI) of Vega Media in the Region of Murcia (Spain), which serves three municipalities. The centre’s psychologist was contacted and informed about the aims and methodology of the study. Following their approval, the Head of the Planning and Programmes Service of the Directorate General for Women and Gender Diversity of the Autonomous Community of the Region of Murcia was contacted to request authorisation for data collection.

Ethical approval was subsequently obtained from the Clinical Research Ethics Committee (CEIC) of the University of Murcia (CEIC-1986/2021). After obtaining all necessary authorizations, data collection was carried out at the CAVI with the collaboration of the center’s psychologist.

Participants were informed about the objectives and procedures of the study and invited to participate voluntarily. Those who agreed signed an informed consent form. Data was collected through individual interviews during which sociodemographic information was recorded and questionnaires were administered. Participants received instructions on how to complete the instruments. The assessment process required approximately 30 min per participant.

The study protocol was preregistered on the Open Science Framework (OSF) (https://osf.io/93cgu/?view_only=86dd9f891761498fb7db11f35edcb315, accessed on 13 April 2026).

5.5. Data Analysis

Data analyses included frequency analysis, descriptive statistics (means and standard deviations) for sociodemographic variables, independent samples t-tests, Pearson correlation coefficients, and multiple linear regression analyses using the enter method.

In addition, tests of normality were conducted using the Shapiro–Wilk statistic. The results of these tests indicated the suitability of applying parametric analyses in the present study.

Missing data were handled using available-case analysis, and variations in sample size across analyses are reported where applicable.

All analyses were conducted in accordance with the exploratory and correlational design of the study, aimed at examining associations between variables rather than establishing causal relationships.

Statistical analyses were performed using IBM SPSS Statistics (version 23.0).

6. Conclusions

This study contributes to the understanding of gender-based violence in rural contexts by describing the sociodemographic profile, mental health indicators, and perceived risk of future violence among women attending a specialised care centre in small municipalities.

The findings suggest that participants presented moderate levels of depression, anxiety, and stress symptoms, together with predominantly low levels of perceived risk of future violence. No statistically significant associations were observed between age, clinical symptomatology, and risk perception. Women without children tended to report higher levels of psychological symptomatology and perceived risk compared to those with children, although these findings should be interpreted with caution given the sample size. Furthermore, the type of abuse showed an exploratory association with perceived risk of future violence, with the set of abuse indicators explaining 21% of its variance; however, this result should be considered preliminary.

Although the sample size limits the generalizability of the findings, the results provide preliminary insights into the relevance of contextual and sociodemographic variables in rural environments when assessing mental health and risk perception in women victims of gender-based violence. These findings may help to inform future research and the development of context-sensitive assessment and intervention strategies in specialized care settings.